1. Field of the Invention
The present invention relates to an area estimation system and an area estimation method.
2. Description of the Related Art
In recent years, users have been provided with many services relating to places where the users currently are.
For example, there is a service which allows users to share, as presence information, information on locations where they currently are. Another example of the services is a push-type information delivery which provides advertisements and information to a user who is staying at a certain location.
In order to provide such services relating to a user's location, it is necessary to successively know where the user currently is.
A location estimation apparatus has been known as a method for estimating a user's location. In a space equipped with a wireless LAN, the location estimation apparatus learns in advance an RSSI (Received Signal Strength Indicator) for indicating a strength of a radio signal which a wireless LAN device or the like possessed by a user receives from each access point in a wireless LAN network, and then estimates the user's location by using the learning result (refer to Gakushu Kino wo Mochiita Location Kenshutu Houhou no Kentou (Study on a method for detecting a location by using a learning function) (NTT, Tomoaki Ogawa, Shuichi Yoshino and Masashi Shimizu, the Institute of Electronics, Information and Communication Engineers, Shingakugihou NS2002-79); and International Patent Publication No. W02004-008795).
Moreover, another location estimation apparatus has been known which estimates the location of a wireless LAN device possessed by a user by using the RSSI measured at a wireless LAN base station, and which thus determines the user's location.
However, the location estimation apparatuses of the foregoing conventional techniques have the following problems.
Specifically, the location estimation apparatuses of the foregoing conventional techniques operate without considering the variation in the RSSI caused by the movement of a user and the like. In General, the RSSI varies due to the movements or the like of a structural object, furniture, people and things.
Since the structural object, the furniture and the like are hardly moved, the movements thereof do not cause the RSSI to vary from moment to moment.
On the other hand, since the people and the things frequently move, the movements thereof cause the RSSI to widely vary from moment to moment.
For this reason, it is difficult to estimate an accurate location, even if the RSSI at each area is learnt in advance as is the case with the foregoing conventional techniques. This is because there is a high possibility that the RSSI measured at an estimation time widely varies due to the movements of the people and the things.
The above-described literatures also describe a fact that the longer the distance between the base station and the location of a target for estimation, the more remarkable such characteristics.
Moreover, the location estimation apparatuses of the foregoing conventional techniques need to learn in advance RSSIs measured at intervals of about one meter for the purpose of estimating the location. This requires huge setting costs for the location estimation.
In addition, in a case where the location is estimated by using the RSSI measured at the wireless LAN base station, the longer the distance between the base station and a target, the more widely the RSSI varies and the more likely an error is to occur.